Get on the bus, or go home? The AI moment in marketing is here
JJ
We've seen this plot in design and marketing before, 30 years ago.
In 1995, I was a young professional who was focused on using digital tools that most of the marketing and design establishment didn't take seriously. I was the "computer guy" on teams that had only just, grudgingly, finished trading their markers and paste up boards for computers. That transition had already been uncomfortable. Plenty of art directors never forgave the mouse. But it didn't break the industry, it just changed the tools in people's hands.
Then the internet emerged. It wasn't just a nerdy tool, but a new communication platform that demanded marketing leadership take it seriously, structurally, immediately. And leadership, by and large, resisted.
I remember what that resistance looked and sounded like from the inside, because I was on the receiving end. I was called ugly names to my face by colleagues who thought I was a novelty act, a hobbyist playing with toys while the "real" design work of marketing kept happening the way it always had, on paper. Committees formed. Pilot programs launched and died. Leadership issued statements about "monitoring the space" with subsequent lunch-and-learns and modest levels of paid training. That was the shape denial took: not silence, but theatrical and symbolic performance.
It didn't last. Within two years, the people mocking me were gone or quieted, and I was running the department. It wasn't because I'd won some political battle, but because in 1995 and 1996, mandates for web-centric marketing started coming down from the very top of corporations and agencies. Not from the middle. The middle was still in open rebellion. The mandate came from the C-suite and the boardroom, and it arrived as an ultimatum: transform, or leave. Tens of thousands of marketers who resisted the internet transition were shown the door.
I believe we are standing in a nearly identical moment right now, except this time the wave of change is bigger, faster, and touches more of what marketing actually does.
The resistance looks mostly the same
Walk into most corporate marketing departments or agency teams today and you'll find a familiar pattern dressed in 2026 clothes. There's an AI tool bolted onto a legacy process so someone can put a checkmark next to "AI-enabled" in a deck. "I use AI" is almost always followed by "ChatGPT." There's a lot of talk about "responsible adoption" that functions mostly as a stalling tactic. The resistance isn't coming from the top, and often it isn't even coming from the bottom; it's coming from the middle layer of management that has the most to lose and the least incentive to move fast.
The data backs up what I'm seeing anecdotally. Gartner's February 2026 research found that 65% of CMOs expect AI to dramatically transform their role within two years, but only 32% believe their own skill set needs significant updating. That gap between what marketing leaders expect from AI and what they're willing to change about themselves is the same gap I watched paralyze marketing departments in 1995. It's denial wearing a survey response. And it's not just self-assessment: only 15% of CEOs currently consider their marketing leader to be AI-savvy, according to the same research. Gartner's 2026 CMO Spend Survey found a similar split: 70% of marketing leaders name AI leadership as a critical goal, but only 30% report their organization has the maturity to actually scale it.
This is not a technology gap. It's a leadership gap. And leadership gaps get closed one of two ways: the people in the gap grow into it, or they get replaced by people who already have.
The purge is already starting, and it's starting at the top of the industry first
Here's where the pattern from the internet era is repeating almost exactly, and where I think a lot of marketing and creative leaders are underestimating how fast this will move once it starts in earnest.
The holding companies, the biggest and most legacy-bound organizations in our industry, are already restructuring around AI, not experimenting with it at the margins. WPP's headcount fell from roughly 108,000 to under 99,000 in a single year as part of an AI-driven efficiency overhaul. Omnicom and IPG shed more than 23,000 combined positions in their AI-accelerated merger. Forrester originally projected agencies would lose about 7.5% of jobs to AI between 2024 and 2032, then revised that forecast to 15% of agency jobs eliminated in 2026 alone, compressing what they thought would take eight years into one.
And it isn't only the laggards cutting. Publicis grew revenue faster than its holding-company peers in the same period specifically because of AI integration, with 85% of its client-facing staff now using its internal AI platform, and it still trimmed headcount doing it. Growing and cutting at the same time isn't a contradiction right now. It's the actual playbook.
This is the part of the 1995 story that I think people forget: the purge didn't happen because the internet made marketing obsolete. It happened because leadership decided it was faster and cheaper to replace people who wouldn't change than to wait for them to change on their own. We can expect the same math to play out here, presented the same way it was to me back then: as inevitable, and as already decided by the time most people hear about it.
The signals are no longer theoretical
For the last couple of years, this has mostly been a conversation about the future. That's changed for me in the last few months, and not in a subtle way.
I'm seeing it in job postings: a real, visible shift in what companies are asking marketing leadership candidates to have already done with AI, not just what they're "interested in exploring." I'm hearing it from recruiters, who are fielding briefs from clients that read less like "nice to have AI experience" and more like a hard requirement. And I'm hearing it directly in the questions coming in about my Brand Intelligence product, not "what is this?" but "how fast can this be running inside our organization?" and "can this replace headcount we're already planning to eliminate?" Those are not the questions of an industry still deciding whether AI transformation is coming. Those are the questions of an industry that has already decided and is now shopping for how to execute.
The broader numbers back up what I'm hearing directly. AI became the single most-cited reason for U.S. job cuts for multiple consecutive months in 2026, and Forrester's own workforce forecast projects roughly 92,000 marketing-adjacent roles in North America undergoing formal restructuring by the end of this year, with the majority of those eliminated outright rather than redefined.
Get on the bus?
I don't say any of this to be alarmist, and I don't say it because I think AI should replace the human judgment, taste, and strategic thinking that make marketing and brand work valuable in the first place. I've built my own AI approach and product on the opposite premise: that AI should extend human creative leadership, not eliminate it. But extending human leadership only works if that leadership fully embraces the tools and adopts a transformation mindset.
In 1995, the marketers who thrived weren't the ones with the most experience. They were the ones willing to be beginners again in public, at a moment when being a beginner in public was genuinely uncomfortable. I was mocked for it. Then, within a year, I wasn't.
The bus is here again. It's bigger this time, it's moving faster, and it is not going to wait at the curb while committees deliberate. Marketing and creative leaders have a real window right now to get on it, to actually rebuild how their teams work, not just bolt an LLM onto the same old process. That window will close the same way it did last time: quietly, then all at once, with a mandate from the top and very little patience for anyone still standing at the curb.